{"title":"Location-Aware mmWave Beamforming Based on Kernel Trick","authors":"Bin Yang, Li Chen, Weidong Wang","doi":"10.1109/WCSP.2019.8928119","DOIUrl":null,"url":null,"abstract":"Location-aware beamforming has been proposed as a potential approach for fast beam selection in millimeter wave (mmWave) multi-input multi-output (MIMO) system. In this paper, we introduce kernel trick to utilize the feature that channel parameters vary continuously with the changing of location. We first formulate the beamforming problem as a classification problem, and we use a multi-class support vector machine to solve this problem. Then, we propose a custom kernel function based on the metric of maximizing the signal-to-noise ratio. Simulation results show that the proposed method performs well in beamforming and it outperforms other commonly used kernel functions.","PeriodicalId":108635,"journal":{"name":"2019 11th International Conference on Wireless Communications and Signal Processing (WCSP)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 11th International Conference on Wireless Communications and Signal Processing (WCSP)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/WCSP.2019.8928119","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Location-aware beamforming has been proposed as a potential approach for fast beam selection in millimeter wave (mmWave) multi-input multi-output (MIMO) system. In this paper, we introduce kernel trick to utilize the feature that channel parameters vary continuously with the changing of location. We first formulate the beamforming problem as a classification problem, and we use a multi-class support vector machine to solve this problem. Then, we propose a custom kernel function based on the metric of maximizing the signal-to-noise ratio. Simulation results show that the proposed method performs well in beamforming and it outperforms other commonly used kernel functions.